On Thu, Nov 12, 2020 at 10:04:17AM +0100, Arnaud POULIQUEN wrote:
>
>
> On 11/11/20 1:37 AM, Mathieu Poirier wrote:
> > On Tue, 10 Nov 2020 at 11:18, Arnaud POULIQUEN <arnaud.pouliquen@xxxxxx> wrote:
> >>
> >> Hi Mathieu, Guennadi,
> >>
> >> On 11/9/20 6:55 PM, Mathieu Poirier wrote:
> >>> On Mon, Nov 09, 2020 at 11:20:24AM +0100, Guennadi Liakhovetski wrote:
> >>>> Hi Arnaud,
> >>>>
> >>>> On Mon, Nov 09, 2020 at 09:48:37AM +0100, Arnaud POULIQUEN wrote:
> >>>>> Hi Guennadi, Mathieu,
> >>>>>
> >>>>> On 11/6/20 6:53 PM, Mathieu Poirier wrote:
> >>>>>> On Fri, Nov 06, 2020 at 03:00:28PM +0100, Guennadi Liakhovetski wrote:
> >>>>>>> On Fri, Nov 06, 2020 at 02:15:45PM +0100, Guennadi Liakhovetski wrote:
> >>>>>>>> Hi Mathieu, Arnaud,
> >>>>>>>>
> >>>>>>>> On Thu, Nov 05, 2020 at 03:50:28PM -0700, Mathieu Poirier wrote:
> >>>>>>>>> From: Arnaud Pouliquen <arnaud.pouliquen@xxxxxx>
> >>>>>>>>>
> >>>>>>>>> Make the RPMSG name service announcement a stand alone driver so that it
> >>>>>>>>> can be reused by other subsystems. It is also the first step in making the
> >>>>>>>>> functionatlity transport independent, i.e that is not tied to virtIO.
> >>>>>>>>
> >>>>>>>> Sorry, I just realised that my testing was incomplete. I haven't tested
> >>>>>>>> automatic module loading and indeed it doesn't work. If rpmsg_ns is loaded
> >>>>>>>> it probes and it's working, but if it isn't loaded and instead the rpmsg
> >>>>>>>> bus driver is probed (e.g. virtio_rpmsg_bus), calling
> >>>>>>>> rpmsg_ns_register_device() to create a new rpmsg_ns device doesn't cause
> >>>>>>>> rpmsg_ns to be loaded.
> >>>>>>>
> >>>>>>> A simple fix for that is using MODULE_ALIAS("rpmsg:rpmsg_ns"); in rpmsg_ns.c
> >>>>>>> but that alone doesn't fix the problem completely - the module does load then
> >>>>>>> but not quickly enough, the NS announcement from the host / remote arrives
> >>>>>>> before rpmsg_ns has properly registered. I think the best solution would be
> >>>>>>> to link rpmsg_ns.c together with rpmsg_core.c. You'll probably want to keep
> >>>>>>> the module name, so you could rename them to just core.c and ns.c.
> >>>>>>
> >>>>>> I'm pretty sure it is because virtio_device_ready() in rpmsg_probe() is called
> >>>>>> before the kernel has finished loading the name space driver. There has to be
> >>>>>> a way to prevent that from happening - I will investigate further.
> >>>>>
> >>>>> Right, no dependency is set so the rpmsg_ns driver is never probed...
> >>>>> And name service announcement messages are dropped if the service is not present.
> >>>>
> >>>> The mentioned change
> >>>>
> >>>> -MODULE_ALIAS("rpmsg_ns");
> >>>> +MODULE_ALIAS("rpmsg:rpmsg_ns");
> >>>
> >>> Yes, I'm good with that part.
> >>>
> >>>>
> >>>> is actually a compulsory fix, without it the driver doesn't even get loaded when
> >>>> a device id registered, using rpmsg_ns_register_device(). So this has to be done
> >>>> as a minimum *if* we keep RPNsg NS as a separate kernel module. However, that
> >>>> still doesn't fix the problem relyably because of timing. I've merged both the
> >>>> RPMsg core and NS into a single module, which fixed the issue for me. I'm
> >>>> appending a patch to this email, but since it's a "fixup" please, feel free to
> >>>> roll it into the original work. But thinking about it, even linking modules
> >>>> together doesn't guarantee the order. I think rpmsg_ns_register_device() should
> >>>> actually actively wait for NS device probing to finish - successfully or not.
> >>>> I can add a complete() / wait_for_completion() pair to the process if you like.
> >>>>
> >>>
> >>> Working with a completion is the kind of thing I had in mind. But I would still
> >>> like to keep the drivers separate and that's the part I need to think about.
> >>
> >> I reproduce the problem: the rpmsg_ns might not be probed on first message reception.
> >> What makes the fix not simple is that the virtio forces the virtio status to ready
> >> after the probe of the virtio unit [1].
> >> Set this status tiggs the remote processor first messages.
> >>
> >> [1]https://elixir.bootlin.com/linux/latest/source/drivers/virtio/virtio.c#L253
> >>
> >> Guennadi: I'm not sure that your patch will solve the problem , look like it just reduces the
> >> delay between the rpmsg_virtio and the rpmsg_ns probe (the module loading time is saved)
> >>
> >> Based on my observations, I can see two alternatives.
> >> - rpmsg_ns.c is no longer an rpmsg driver but a kind of function library to manage a generic name service.
> >
> > That option joins Guennadi's vision - I think he just expressed it in
> > a different way. The more I think about it, the more I find that
> > option appealing. With the code separation already achieved in this
> > patchset it wouldn't be hard to implement.
>
> Right, similar to Guennadi's version, if we want to keep it simpler this is
> probably the preferred option.
> From my point of view the main requierement is that the ns announcement service
> is generic.
>
>
> >
> >> - we implement a completion as proposed by Mathieu.
> >>
> >> I tried this second solution based on the component bind mechanism.
> >> I added the patch at the end of the mail (the patch is a POC, so not ready for upstream).
> >> Maybe something simpler is possible. I'm just keeping in mind that we may have to add similar
> >> services in the future.
> >>
> >
> > Wasn't familiar with the "component" infrastructure - I suppose you
> > stumbled on it while working on sound drivers. I have to spend more
> > time looking at it.
>
> Used in DRM framework mainly, i implemented this in my RFC[1] concerning the
> refactoring of the rproc_virtio in a platform driver. The idea was to ensure
> that all rproc sub-devices are registered before starting the remote processor.
>
> [1]https://lkml.org/lkml/2020/4/16/1817
>
> The principle it to attach child components to a master component, this
> relationship allows to synchronize all using component_master_add_with_match
> and component_bind_all after the drivers probing step.
> The drawback of this solution is that it make code more complex.
>
> > But if you have time and want to spin off a new
> > revision that implements the library concept, I'll invest time on that
> > instead.
>
> Time is always a major issue :)
> No time this week, but i will try to send patches next week.
I agree, I'm running short on it too. I thought what you pointed out with
find_module() [1] was quite interesting. I tried it on my side but used
request_module() rather than request_module_nowait() - because we do want to
wait for the namespace driver to be loaded before moving forward. Unfortunatly
the system hung on me... I still don't know why, I will have to investigate
further.
Regarding your patch on components, I am now up to speed with the concept.
That too is interesting but likely an overkill for the current
situation. Right now we have a single driver to deal with and I think we should
keep things as simple as possible.
I'll talk to you guys on Monday,
Mathieu
[1]. https://elixir.bootlin.com/linux/latest/source/drivers/gpu/drm/drm_fb_helper.c#L2274
>
> Regards,
> Arnaud
>
> >
> >> Regards,
> >> Arnaud
> >>
> >> From f2de77027f4a3836f8bf46aa257e5592af6529b7 Mon Sep 17 00:00:00 2001
> >> From: Arnaud Pouliquen <arnaud.pouliquen@xxxxxx>
> >> Date: Tue, 10 Nov 2020 18:39:29 +0100
> >> Subject: [PATCH] rpmsg_ns: add synchronization based on component mechanism
> >>
> >> Implement the component bind mechanism to ensure that the rpmsg virtio bus
> >> driver are probed before treating the first RPMsg.
> >>
> >> Signed-off-by: Arnaud Pouliquen <arnaud.pouliquen@xxxxxx>
> >> ---
> >> drivers/rpmsg/rpmsg_ns.c | 26 ++++++++++++-
> >> drivers/rpmsg/virtio_rpmsg_bus.c | 65 ++++++++++++++++++++++++++++++++
> >> 2 files changed, 89 insertions(+), 2 deletions(-)
> >>
> >> diff --git a/drivers/rpmsg/rpmsg_ns.c b/drivers/rpmsg/rpmsg_ns.c
> >> index 5bda7cb44618..057e5d1d29a0 100644
> >> --- a/drivers/rpmsg/rpmsg_ns.c
> >> +++ b/drivers/rpmsg/rpmsg_ns.c
> >> @@ -2,6 +2,7 @@
> >> /*
> >> * Copyright (C) STMicroelectronics 2020 - All Rights Reserved
> >> */
> >> +#include <linux/component.h>
> >> #include <linux/device.h>
> >> #include <linux/kernel.h>
> >> #include <linux/module.h>
> >> @@ -55,6 +56,24 @@ static int rpmsg_ns_cb(struct rpmsg_device *rpdev, void *data, int len,
> >> return 0;
> >> }
> >>
> >> +static int rpmsg_ns_bind(struct device *dev, struct device *master, void *data)
> >> +{
> >> + dev_info(dev, "rpmsg ns bound\n");
> >> +
> >> + return 0;
> >> +}
> >> +
> >> +static void rpmsg_ns_unbind(struct device *dev, struct device *master,
> >> + void *data)
> >> +{
> >> + dev_info(dev, "rpmsg ns unbound\n");
> >> +}
> >> +
> >> +static const struct component_ops rpmsg_ns_ops = {
> >> + .bind = rpmsg_ns_bind,
> >> + .unbind = rpmsg_ns_unbind,
> >> +};
> >> +
> >> static int rpmsg_ns_probe(struct rpmsg_device *rpdev)
> >> {
> >> struct rpmsg_endpoint *ns_ept;
> >> @@ -63,6 +82,7 @@ static int rpmsg_ns_probe(struct rpmsg_device *rpdev)
> >> .dst = RPMSG_NS_ADDR,
> >> .name = "name_service",
> >> };
> >> + int ret;
> >>
> >> /*
> >> * Create the NS announcement service endpoint associated to the RPMsg
> >> @@ -76,7 +96,9 @@ static int rpmsg_ns_probe(struct rpmsg_device *rpdev)
> >> }
> >> rpdev->ept = ns_ept;
> >>
> >> - return 0;
> >> + ret = component_add(&rpdev->dev, &rpmsg_ns_ops);
> >> +
> >> + return ret;
> >> }
> >>
> >> static struct rpmsg_driver rpmsg_ns_driver = {
> >> @@ -104,5 +126,5 @@ module_exit(rpmsg_ns_exit);
> >>
> >> MODULE_DESCRIPTION("Name service announcement rpmsg Driver");
> >> MODULE_AUTHOR("Arnaud Pouliquen <arnaud.pouliquen@xxxxxx>");
> >> -MODULE_ALIAS("rpmsg_ns");
> >> +MODULE_ALIAS("rpmsg:rpmsg_ns");
> >> MODULE_LICENSE("GPL v2");
> >> diff --git a/drivers/rpmsg/virtio_rpmsg_bus.c b/drivers/rpmsg/virtio_rpmsg_bus.c
> >> index 30ef4a5de4ed..c28aac1295fa 100644
> >> --- a/drivers/rpmsg/virtio_rpmsg_bus.c
> >> +++ b/drivers/rpmsg/virtio_rpmsg_bus.c
> >> @@ -11,6 +11,7 @@
> >>
> >> #define pr_fmt(fmt) "%s: " fmt, __func__
> >>
> >> +#include <linux/component.h>
> >> #include <linux/dma-mapping.h>
> >> #include <linux/idr.h>
> >> #include <linux/jiffies.h>
> >> @@ -67,11 +68,16 @@ struct virtproc_info {
> >> struct mutex endpoints_lock;
> >> wait_queue_head_t sendq;
> >> atomic_t sleepers;
> >> + struct component_match *match;
> >> + struct completion completed;
> >> + int bind_status;
> >> };
> >>
> >> /* The feature bitmap for virtio rpmsg */
> >> #define VIRTIO_RPMSG_F_NS 0 /* RP supports name service notifications */
> >>
> >> +#define BIND_TIMEOUT_MS 1000
> >> +
> >> /**
> >> * struct rpmsg_hdr - common header for all rpmsg messages
> >> * @src: source address
> >> @@ -768,6 +774,17 @@ static void rpmsg_recv_done(struct virtqueue *rvq)
> >> unsigned int len, msgs_received = 0;
> >> int err;
> >>
> >> + /* Wait for all children to be bound */
> >> + if (vrp->bind_status) {
> >> + dev_dbg(dev, "cwait bind\n");
> >> + if (!wait_for_completion_timeout(&vrp->completed,
> >> + msecs_to_jiffies(BIND_TIMEOUT_MS)))
> >> + dev_err(dev, "child device(s) binding timeout\n");
> >> +
> >> + if (vrp->bind_status)
> >> + dev_err(dev, "failed to bind RPMsg sub device(s)\n");
> >> + }
> >> +
> >> msg = virtqueue_get_buf(rvq, &len);
> >> if (!msg) {
> >> dev_err(dev, "uhm, incoming signal, but no used buffer ?\n");
> >> @@ -808,6 +825,39 @@ static void rpmsg_xmit_done(struct virtqueue *svq)
> >> wake_up_interruptible(&vrp->sendq);
> >> }
> >>
> >> +static int virtio_rpmsg_compare(struct device *dev, void *data)
> >> +{
> >> + return dev == data;
> >> +}
> >> +
> >> +static void virtio_rpmsg_unbind(struct device *dev)
> >> +{
> >> + /* undbind all child components */
> >> + component_unbind_all(dev, NULL);
> >> +}
> >> +
> >> +static int virtio_rpmsg_bind(struct device *dev)
> >> +{
> >> + struct virtio_device *vdev = dev_to_virtio(dev);
> >> + struct virtproc_info *vrp = vdev->priv;
> >> +
> >> + dev_dbg(dev, "Bind virtio rpmsg sub devices\n");
> >> +
> >> + vdev = container_of(dev, struct virtio_device, dev);
> >> + vrp->bind_status = component_bind_all(dev, NULL);
> >> + if (vrp->bind_status)
> >> + dev_err(dev, "bind virtio rpmsg failed\n");
> >> +
> >> + complete(&vrp->completed);
> >> +
> >> + return vrp->bind_status;
> >> +}
> >> +
> >> +static const struct component_master_ops virtio_rpmsg_cmp_ops = {
> >> + .bind = virtio_rpmsg_bind,
> >> + .unbind = virtio_rpmsg_unbind,
> >> +};
> >> +
> >> static int rpmsg_probe(struct virtio_device *vdev)
> >> {
> >> vq_callback_t *vq_cbs[] = { rpmsg_recv_done, rpmsg_xmit_done };
> >> @@ -892,6 +942,7 @@ static int rpmsg_probe(struct virtio_device *vdev)
> >> /* if supported by the remote processor, enable the name service */
> >> if (virtio_has_feature(vdev, VIRTIO_RPMSG_F_NS)) {
> >> vch = kzalloc(sizeof(*vch), GFP_KERNEL);
> >> +
> >> if (!vch) {
> >> err = -ENOMEM;
> >> goto free_coherent;
> >> @@ -911,6 +962,20 @@ static int rpmsg_probe(struct virtio_device *vdev)
> >> err = rpmsg_ns_register_device(rpdev_ns);
> >> if (err)
> >> goto free_coherent;
> >> + /* register a component associated to the virtio platform */
> >> + component_match_add_release(&vdev->dev, &vrp->match,
> >> + NULL, virtio_rpmsg_compare,
> >> + &rpdev_ns->dev);
> >> +
> >> + vrp->bind_status = -ENXIO;
> >> + init_completion(&vrp->completed);
> >> + err = component_master_add_with_match(&vdev->dev,
> >> + &virtio_rpmsg_cmp_ops,
> >> + vrp->match);
> >> + if (err) {
> >> + dev_err(&vdev->dev, "failed to bind virtio rpmsg\n");
> >> + goto free_coherent;
> >> + }
> >> }
> >>
> >> /*
> >> --
> >> 2.17.1
> >>
> >>
> >>
> >>
